Abstract The proposed research addresses two fundamental problems in population research on the Alzheimer's disease (AD): (1) sex disparities in cognitive decline with aging, and (2) interplay of social and biological pathways that generate and sustain AD disparities. Our proposal breaks new ground with an innovative life- course research design that conjoins prospective cohort data from five large-scale NIH population-based longitudinal studies that collectively cover the adult life span. We will conduct integrative data analysis in response to an increased demand for replicability of scientific findings across independent studies and make efficient use of extant resources in the pursuit of a cumulative science. We will apply novel statistical methods to model longitudinal sex differences in age trajectories of cognitive functioning from young to late adulthood and incidence of AD in old age. We will examine social and biological mechanisms influencing cognitive trajectories and sex differences. Tested mechanisms will include social stress related to socioeconomic status (SES) and social relationships, and biomarkers of cardiovascular, metabolic, neuroendocrine, and immune pathways implicated in AD. We will also test and integrate new biological data on immune function.